Topical preparations for preventing stretch marks in pregnancy (Review) The Cochrane Library

Transcription

Topical preparations for preventing stretch marks in
pregnancy (Review)
Brennan M, Young G, Devane D
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2012, Issue 11
http://www.thecochranelibrary.com
Topical preparations for preventing stretch marks in pregnancy (Review)
Copyright © 2012 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 1.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 2.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGEMENTS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.1. Comparison 1 Topical preparations with active ingredients compared with placebo or no treatment, Outcome
1 Presence of stretch marks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.2. Comparison 1 Topical preparations with active ingredients compared with placebo or no treatment, Outcome
2 Severity of stretch marks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.1. Comparison 2 Topical preparations with active ingredients compared with other topical preparations with
active ingredient, Outcome 1 Presence of stretch marks. . . . . . . . . . . . . . . . . . . .
Analysis 2.2. Comparison 2 Topical preparations with active ingredients compared with other topical preparations with
active ingredient, Outcome 2 Severity of stretch marks. . . . . . . . . . . . . . . . . . . .
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . .
NOTES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
INDEX TERMS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
1
2
2
4
4
8
9
10
12
14
14
14
17
29
30
31
32
32
33
33
33
34
34
34
34
34
i
[Intervention Review]
Topical preparations for preventing stretch marks in
pregnancy
Miriam Brennan1 , Gavin Young2 , Declan Devane1
1 School of Nursing and Midwifery, National University of Ireland Galway, Galway, Ireland. 2 Temple Sowerby Medical Practice, Penrith,
UK
Contact address: Gavin Young, Temple Sowerby Medical Practice, Linden Park, Temple Sowerby, Penrith, Cumbria, CA10 1RW, UK.
gavinlyoung@btinternet.com.
Editorial group: Cochrane Pregnancy and Childbirth Group.
Publication status and date: New search for studies and content updated (conclusions changed), published in Issue 11, 2012.
Review content assessed as up-to-date: 6 March 2012.
Citation: Brennan M, Young G, Devane D. Topical preparations for preventing stretch marks in pregnancy. Cochrane Database of
Systematic Reviews 2012, Issue 11. Art. No.: CD000066. DOI: 10.1002/14651858.CD000066.pub2.
ABSTRACT
Background
Striae gravidarum (stretch marks developing during pregnancy) occur in 50% to 90% of women. They appear as red or purple lines
or streaks that fade slowly to leave pale lines or marks on the skin. The abdomen, breasts and thighs are commonly affected. The exact
cause of stretch marks is unclear and no preparation has yet been shown to be effective in preventing the development of stretch marks.
They are a source of significant anxiety for women, impacting on their quality of life.
Objectives
To assess the effects of topical preparations on the prevention of stretch marks in pregnancy.
Search methods
We searched the Cochrane Pregnancy and Childbirth Group’s Trials Register (31 October 2011) and reference lists of retrieved reports.
Selection criteria
We included randomised controlled trials and quasi-randomised controlled trials comparing topical preparations (with active ingredients) with other topical preparations (with active ingredients), with a placebo (that is, preparations without active ingredients) or with
no treatment for the prevention of stretch marks in pregnant women.
Data collection and analysis
Three review authors independently assessed trial eligibility and trial quality, and extracted data. Data were checked for accuracy. The
primary outcome was the presence of stretch marks and the secondary outcome was the severity of stretch marks.
Main results
We included six trials involving 800 women. Of the six trials, we judged the risk of bias for three as ’low risk’ for random sequence
generation, blinding of participants and personnel, blinding of outcome assessment, completeness of outcome data and selective
reporting.
1
There was no statistically significant average difference in the development of stretch marks in women who received topical preparations
with active ingredients compared to women who received a placebo or no treatment (average risk ratio (RR) 0.74; 95% confidence
interval (CI) 0.53 to 1.03; five trials, 474 women; random-effects model, Tau² = 0.09, I² = 65%) (Analysis 1.1).
Results were consistent with the main effects when we performed a sensitivity analysis excluding studies judged to be at high risk of
bias for random sequence generation, allocation concealment or more than 20% missing data for a given outcome (average RR 0.81;
95% CI 0.60 to 1.10; four trials, 424 women; random-effects model, Tau² = 0.05, I² = 57%).
The was no statistically significant average mean difference in the severity of stretch marks (standardised mean difference (SMD) -0.31;
95% CI -1.06 to 0.44; two trials, 255 women; Tau² = 0.26, I² = 87%).
There was no statistically significant difference in the development of stretch marks in women who received topical preparations with
active ingredients compared to women who received other topical preparations with active ingredients (average RR 0.51; 95% CI 0.16
to 1.60; two trials, 305 women; Tau² = 0.53, I² = 74%). There was no statistically significant difference in the severity of stretch marks
(mean difference (MD) -0.20; 95% CI -0.53 to 0.13; one trial, 206 women; heterogeneity not applicable).
Authors’ conclusions
We found no high-quality evidence to support the use of any of the topical preparations in the prevention of stretch marks during
pregnancy. There is a clear need for robust, methodologically rigorous randomised trials involving larger sample sizes to evaluate
the effects of topical preparations on the development of stretch marks in pregnancy. In addition, it is important that preparations
commonly used by women to prevent and treat stretch marks are evaluated within the context of robust, methodologically rigorous
and adequately powered randomised trials.
PLAIN LANGUAGE SUMMARY
Topical preparations for preventing stretch marks in pregnancy
Stretch marks commonly develop during pregnancy, particularly in the third trimester. They affect 50% to 90% of women. They appear
as red lines or streaks that fade slowly after the pregnancy to leave pale lines on the skin. The abdomen, breasts and thighs are most
often affected. They do not disappear entirely, therefore any treatment which prevents them would be welcomed by many women. In
this review, we identified randomised controlled trials and quasi-randomised controlled trials that compared topical creams, lotions
and ointments containing active ingredients with placebo or no treatment, and topical preparations with active ingredients versus other
topical preparations.
We included six trials (involving 800 women) in this review. We found that the application of a skin preparation to the areas affected
by stretch marks during pregnancy did not prevent the development of stretch marks in the women during pregnancy. Only three trials
(involving 461 women) looked at the severity of the stretch marks and did not show a clear difference. The preparations used included
Alphastria, Trofolastin, Verum, olive oil and cocoa butter, which all contain vitamin E; Alphastria and Verum also have hyaluronic
acid. Of the six trials, we judged three to be at low risk of bias. All trials were relatively small, with four of the six trials each including
less than 100 women. The trials were also different in terms of when the women first started to use the topical applications, ranging
from the first trimester to the first 20 weeks.
Description of the condition
BACKGROUND
The following review is an update of the review ‘Creams for preventing stretch marks in pregnancy’ (Young 1996).
Striae distensae (stretch marks), or striae gravidarum as they are
known in pregnancy (Cunningham 2010), are considered to be
the most common connective tissue change in pregnancy (Lawley
1999). Rates of occurrence of striae gravidarum vary (Salter 2006),
2
with reported rates ranging between 50% and 90% (Osman
2007). In primiparous women incidences of 52% (Atwal 2006),
61% (Osman 2007) and 87.7% (Ghasemi 2007) have been reported, while a rate of 71.1% was found in a study involving
both primigravidae and multigravidae (Muzaffar 1998). Striae
gravidarum seem to affect all racial groups (Buchanan 2010). Although once considered to be more common in white than in
black or Asian women (Wong 1984; Wong 1989), more recently
non-white women were seen to be at greater risk (Chang 2004).
Striae gravidarum are common during the first pregnancy (Salter
2006) and usually present during the third trimester (Atwal 2006;
Cunningham 2010). However, there have been reports in women
under 24 weeks’ gestation and of women first developing them in
a second pregnancy (Chang 2004).
Striae have been defined as ’visible linear scars’ (Burrows 2004:
46.6) that have evolved through recognised stages (Kang 1996)
similar to the stages of tissue healing (Kang 1998; Salter 2006)
or scar formation (Elson 1990). They manifest as ’reddish slightly
depressed streaks’ (Cunningham 2010: 111) or ’reddish purple linear macules’ (Horn 2007: 947). They often fade gradually (Kang
1996; Kang 1998; Papoutsis 2007; Salter 2006) leaving glistening
(Cunningham 2010), white depressed (Elson 1990) or pale wrinkled lines (Watson 1998) on the skin, from about six months following birth (Murray 2009). These glistening lines are commonly
seen on multiparous women in addition to the reddish striae of
the current pregnancy (Cunningham 2010). These benign skin
changes (Atwal 2006) commonly occur on the abdomen but are
also seen on the breasts and thighs (Cunningham 2010; Horn
2007; Osman 2008; Salter 2006; Thomas 2004), hips and buttocks (Horn 2007; Osman 2008) and groin and axillae (Papoutsis
2007). Striae have been reported as ranging in severity and have
been graded as mild, moderate or severe by some authors (Atwal
2006; Osman 2007; Osman 2008). Atwal 2006: 966 developed
and used a numerical system that captured the severity of striae,
focusing on the number of striae present and the degree of erythema, or redness. A score of zero to three represented no striae
or ’no significant striae’, four to nine was considered ’mild’, 10
to 15 as ’moderate’ and greater than 16 represented ’severe striae’.
Other criteria for assessing the severity of striae gravidarum include degrees of ’scaling, burning or stinging, or pruritus’ (Kang
1996:520).
While attracting much discussion and debate over the years
(Nigam 1989), the exact cause or origin of striae gravidarum
remains in doubt (Ghasemi 2007; Lawley 1999; Osman 2007;
Osman 2008; Wong 1984) and is understood poorly (Burrows
2010), with researchers disagreeing about their histopathological
origins (Zheng 1985). Nevertheless, several risk factors have been
identified. Early researchers attributed the development of striae
to stretching (Wilks, 1861 cited by Poidevin 1959) and the stretch
theory was accepted widely as the cause of striae gravidarum up
until the middle of the last century (Poidevin 1959) when it became evident that other factors such as increased adrenal cortical
activity may be involved (Poidevin 1959).
From his study of 116 primigravid women, Poidevin 1959 concluded that striae development was not solely reliant on stretching and that striae gravidarum should not be referred to as stretch
marks. Poidevin 1959 proposed the existence of a ’striae factor’
for each woman and while not identifying what this ’striae factor’
may be, he found a clear relationship between the reduced glucose
tolerance in pregnancy, a sign of adrenocortical hyperactivity, and
the development of striae. This link between increased adrenocortical hormonal activity and striae gravidarum has been suggested
by others (Liu 1974; McKenzie 1971). Liu 1974 asserts that striae
gravidarum only develop in oestrogen and relaxin primed connective tissue, in response to stretching. Further, increased corticosteroid levels in pregnancy (Venning 1946) are thought to be
a contributing factor. Oestrogen, relaxin and corticosteroids are
thought to promote the formation of a type of mucopolysaccharide ground substance which promotes separation of the collagen
fibrils (Bryant 1968) and the formation of striae gravidarum in
response to stretch (Liu 1974). Collagen is responsible for the tensile strength of the skin (Waugh 2010) and under normal conditions the interfibrillar substance is highly viscous and there is no
slipping or separation of collagen fibrils (Archer 2004). In pregnancy, the collagen mechanism is disrupted and irreversible sliding
and separation of fibres occurs (Archer 2004). Liu 1974’s position
on the development of striae gravidarum is challenged by Shuster
1979, who contends that while the hormones of pregnancy may
alter the collagen fibrils, there is no evidence to support this. Instead, Shuster 1979 suggests that striae are always due to stretching and, furthermore, only occur in immature connective tissue
characterised by a “critical titre of rigid cross-linked collagen and
elastic unlinked collagen” (Shuster 1979: 161), which may be a
factor in the higher risk of striae in younger women identified in
some studies (Atwal 2006; Murphy 1992; Thomas 2004). The
stretching factor is supported by Thomas 2004 who suggest that
the degree of stretch applied is also influential.
Further insight into the pathogenesis of striae is given by Watson
1998 who suggests that the development of striae is related to
changes in the dermal elastic fibres rather than the collagen. They
hypothesised that striae may occur in individuals where there is
a deficiency in ’cutaneous fibrillin’ and can arise in conditions
like pregnancy where there is extra stretching on the skin. The
extra strain or stretching could be sufficient to tear the elastic
fibre network, resulting in the formation of striae (Watson 1998).
Perhaps corticosteroids may also be influential here as they are
thought to weaken the ’dermal elastic fibres’ leading to their tearing
(McKenzie 1971: 774). However, it is far from conclusive, as
Zheng 1985 suggest that striae are scars and are not due to rupture
of the connective tissue in response to stress. They found that the
elastic fibres and collagen arrangement were in keeping with a scar.
Furthermore, they are characterised by absent rete ridges and a
thinning and flattening of the overlying epidermis (Zheng 1985)
and are devoid of sweat glands or hair follicles.
3
While hormonal influences and stress or stretching factors continue to be considered important in the development of striae
(Lawley 1999), other risk factors have been associated with the development of striae gravidarum (Salter 2006). Identified risk factors include family history, race, skin type, birthweight, baseline
body mass index, weight gain and inadequate nutrition (Osman
2007), younger maternal age, increased pregnancy weight gain,
use of corticosteroids and a genetic susceptibility (Papoutsis 2007).
A number of researchers identified younger maternal age as a
risk factor for the development of striae (Atwal 2006; Murphy
1992; Thomas 2004) while others found no association with age
(Ghasemi 2007). Greater weight gain (Atwal 2006; Murphy 1992)
and higher body mass (Thomas 2004) have been identified as significant factors in the development of striae by some researchers
while Chang 2004 indicated that weight gain and changes in
weight during pregnancy were less predictive of the development
of striae than were genetic factors. A personal history of breast or
thigh striae and genetic factors were thought to be the most predictive for the development of striae (Chang 2004). Family history
was also identified by Osman 2007, where women with a family
history of striae gravidarum were more likely to have moderate to
severe striae gravidarum compared to those with no family history.
Finally, a number of researchers have identified a significant relationship between the development of striae gravidarum and an increased infant birthweight (Atwal 2006; Ghasemi 2007; Murphy
1992).
Striae have been a significant anxiety for women since early times
(Salter 2006). They are an aesthetic concern for many women
(Atwal 2006; Chang 2004; Ghasemi 2007; Osman 2007; Osman
2008; Rangel 2001) and can also be a source of stress (Chang 2004;
Mallol 1991; Salter 2006). They may also cause itching (Horn
2007; Lawley 1999; Martius 1973; Muzaffar 1998; Papoutsis
2007; Salter 2006) or a burning sensation (Salter 2006) for some
women. Authors differ in their evaluation of how symptomatic
or not they are; some see them as often symptomatic (Salter
2006) while others report them as usually asymptomatic (Papoutsis
2007).
Description of the intervention and how the
intervention might work
Many writers refer to the challenges of treating striae (Alster 1997;
Elsaie 2009; Papoutsis 2007), while their prevention has attracted
somewhat less attention. Some argue that it may not be possible to
prevent striae (Cunningham 2010). Yet, there are an abundance
of products on the market claiming to prevent striae (Summers
2009). Consequently, over the years women have used many approaches and preparations to either prevent or treat striae gravidarum, and often at great expense (Salter 2006). It appears that
there are no specific treatments for striae (Elsaie 2009; Errickson
1994; Salter 2006) and no preparation has yet been found to be
effective in preventing or healing the lines that remain (Papoutsis
2007). Approaches or preparations used in the prevention and
treatment include topical preparations, lasers or pulsed light (Elsaie
2009). However, only topical preparations are considered safe to
use in pregnancy and the theoretical reasoning for how they are
thought to work include:
• stimulation of fibroblastic activity leading to increased
production of collagen and fibronectin (Brinkhaus 2000; Elsaie
2009);
• increased blood perfusion through massaging of the area
and potential anti-inflammatory effects (Wierrani 1992);
• Increased skin hydration (Elsaie 2009).
Why it is important to do this review
Striae gravidarum affect between 50% and 90% of women
(Osman 2007) during pregnancy and usually remain as silvery scar
lines on the skin. They are an unwanted consequence of pregnancy,
impacting on women’s perception of themselves (Osman 2008)
and their quality of life (Salter 2006), and are thus of significant
concern to women of child bearing age.
There are many unproven products on the market (Burrows 2010)
tried by many women. Consequently, many women incur great
expense (Salter 2006) trying to prevent or treat striae (Osman
2008). It is important, therefore, to systematically assess the evidence on the effectiveness of these creams and preparations in the
prevention of striae. The findings of this review will benefit both
women and healthcare professionals. The review will assist women
to make informed decisions about their choice of treatment to prevent striae gravidarum and inform healthcare practitioners when
advising women on the effectiveness of topical preparation for the
prevention of striae gravidarum.
OBJECTIVES
To assess the effects of topical preparations on the prevention of
stretch marks in pregnancy.
METHODS
Criteria for considering studies for this review
Types of studies
All randomised and quasi-randomised controlled trials comparing topical preparations (with active ingredients) with other topical preparations (with active ingredients), with a placebo (that is,
preparations without active ingredients) or with no treatment.
4
Types of participants
Pregnant women prior to 20 weeks’ gestation, including women
expecting their first or subsequent babies and women experiencing
multiple pregnancies.
Secondary outcomes
1. Severity of stretch marks
Search methods for identification of studies
Types of interventions
For the purpose of this review topical preparations are categorised
as follows.
1.
Creams or lotions with active ingredients
Creams and lotions are defined as emulsions with moisturising and
emollient effects. They can be either be an ’oil-in-water’ or a ’waterin-oil’ emulsion (Hunter 1973). Viscosity determines whether an
emulsion is categorised as a lotion or a cream.
Examples include:
• Trofolastin cream (containing Centella asiatica extract,
alpha tocopherol and collagen-elastin hydrolysates) (Mallol
1991);
• Alphastria cream (containing hyaluronic acid, vitamins A
and E, allantoin and calcium pantothenate) (de Buman 1987);
• cocoa butter lotion (containing cocoa butter and
tocopheryl acetate (vitamin E) (Osman 2008).
2.
Ointments with active ingredients
Ointments are defined as semi-solid preparations and can be of
three types: those that are ’water soluble’, those that ’emulsify with
water’, or those that ’repel water’ (Hunter 1973: 412).
Examples include:
• Verum ointment (containing vitamin E, essential free fatty
acids, panthenol, hyaluronic acid, elastin and menthol)
(Wierrani 1992).
For the purpose of this review, a placebo is a topical preparation
without active ingredients, or ’no treatment’.
Comparisons
1. Topical preparations with active ingredients compared with
placebo or no treatment
2. Topical preparations with active ingredients compared with
other topical preparations with active ingredients
Types of outcome measures
Primary outcomes
1. Presence of stretch marks
Electronic searches
We searched the Cochrane Pregnancy and Childbirth Group’s Trials Register by contacting the Trials Search Co-ordinator (31 October 2011). The Trials Register is maintained by the Trials Search
Co-ordinator and contains trials identified from:
1. quarterly searches of the Cochrane Central Register of
Controlled Trials (CENTRAL);
2. weekly searches of MEDLINE;
3. weekly searches of EMBASE;
4. handsearches of 30 journals and the proceedings of major
conferences;
5. weekly current awareness alerts for a further 44 journals
plus monthly BioMed Central email alerts.
Details of the search strategies for CENTRAL, MEDLINE and
EMBASE, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section
within the editorial information about the Cochrane Pregnancy
and Childbirth Group.
Trials identified through the searching activities described above
are each assigned to a review topic (or topics). The Trials Search
Co-ordinator searches the register for each review using the topic
list rather than keywords.
Searching other resources
We searched the reference lists of all the identified studies and
retrieved one trial (Msika 2002).
We did not apply any language restrictions.
Data collection and analysis
For methods used in previous versions of this review, please see
Appendix 1.
Methods for this update of the review are informed by the Cochrane
Handbook for Systematic Reviews of Interventions (Higgins 2011).
Selection of studies
Three review authors (M Brennan, D Devane, and G Young (MB,
DD and GY) independently assessed all potential studies identified
for inclusion as a result of the search strategy. We would have
resolved any disagreements through discussion but this was not
necessary.
5
Data extraction and management
We designed a form to extract data. For eligible studies, three
review authors (MB, DD and GY) extracted the data using the
agreed form. We resolved discrepancies through discussion. We
contacted authors from two trials (Horace Fletcher for Buchanan
2010; P Msika for Msika, unknown year) for further information
(see notes in Characteristics of included studies). All data were
entered into the Review Manager software (RevMan 2011) and
checked for accuracy by the three review authors (MB, DD and
GY).
Assessment of risk of bias in included studies
Three review authors (MB, DD and GY) independently assessed
the risk of bias for each study using criteria outlined in the Cochrane
Handbook for Systematic Reviews of Interventions (Higgins 2011).
If there had been any discrepancies we would have resolved them
through discussion, but this was not necessary.
(1) Random sequence generation (checking for possible
selection bias)
We described for each included study the method used to generate
the allocation sequence in sufficient detail to allow an assessment
of whether it should produce comparable groups.
We assessed the method as:
• low risk (any truly random process, e.g. random number
table; computer random number generator);
• high risk (any non-random process, e.g. odd or even date
of birth; hospital or clinic record number);
• unclear risk (insufficient information to permit judgment).
(2) Allocation concealment (checking for possible selection
bias)
We described for each included study the method used to conceal
the allocation sequence and determined whether intervention allocation could have been foreseen in advance of or during recruitment, or changed after assignment.
We assessed the methods as:
• low risk (e.g. telephone or central randomisation;
consecutively numbered sealed opaque envelopes);
• high risk (open random allocation; unsealed or non-opaque
envelopes; alternation; date of birth);
(3) Blinding of participants and personnel (checking for
possible performance bias)
We described for each included study the methods used, if any, to
blind study participants and personnel from knowledge of which
intervention a participant received. We considered that studies
were at low risk of bias if they were blinded or if we judged that
the lack of blinding could not have affected the results.
• low risk, high risk or unclear risk for participants;
• low risk, high risk or unclear risk for personnel.
(4) Blinding of outcome assessment (checking for possible
detection bias)
We described for each included study the methods used, if any, to
blind outcome assessors from knowledge of which intervention a
participant received. We considered that studies were at low risk of
bias if they were blinded, or if we judged that the lack of blinding
could not have affected the results.
We assessed the methods as:·
• low risk (no blinding of outcome assessment but the authors
judged that the outcome was not likely to be influenced by this);
• high risk (no blinding of outcome assessment and the
outcome measurement was likely to have been influenced by
this);
• unclear risk (insufficient information to permit judgment;
the study did not address this).
(5) Incomplete outcome data (checking for possible attrition
bias through withdrawals, dropouts, protocol deviations)
We described for each included study, and for each outcome or
class of outcomes, the completeness of data including attrition
and exclusions from the analysis. We stated whether attrition and
exclusions were reported, the numbers included in the analysis
at each stage (compared with the total number of randomised
participants), reasons for attrition or exclusion where reported,
and whether missing data were balanced across groups or were
related to outcomes. Where sufficient information was reported,
or was supplied by the trial authors, we re-included the missing
data in the analyses we undertook.
• low risk (20% or less of missing data);
• high risk (more than 20% of missing data);
• unclear risk (insufficient reporting to permit judgment; the
study did not address this).
(6) Selective reporting (checking for reporting bias)
We investigated the possibility of selective outcome reporting bias
by identifying the outcomes in the study protocol (if available) and
in the methods section of the publication, and by cross-checking
to see if these outcomes were reported in the results section of the
trial publication(s). PubMed and the World Health Organization
(WHO) International Clinical Trials Registry Platform (ICTRP)
(http://www.who.int/ictrp/en/) were searched for the study protocols.
6
• low risk (where it was clear that all of the study’s prespecified outcomes as identified in the study protocol (where
available) and in the method’s section were reported on; that all
expected outcomes of interest to the review were reported on);
• high risk (where it was clear that not all of the study’s prespecified outcomes as identified in the study protocol (where
available) and in the method’s section were reported on; failure
to include a key outcome that would have been expected to have
been included);
randomised trials and individually-randomised trials we will synthesise the relevant information. We will consider it reasonable to
combine the results from both where there is little heterogeneity
between the study designs and where we consider that there is unlikely to be an interaction between the effect of the intervention
and the choice of randomisation unit. We will acknowledge heterogeneity in the unit of randomisation and perform a sensitivity
analysis to investigate the effects of this heterogeneity on the review findings.
Dealing with missing data
(7) Other bias (checking for other biases)
We described for each included study any important concerns we
had about other possible sources of bias. We assessed whether each
study was free of other problems that could put it at risk of bias as
follows:
• low risk (study appeared to be free of bias);
• high risk (had at least one important risk of bias, for
example related to study design);
For included studies, we noted the levels of attrition.
For all outcomes we carried out analyses, as far as possible, on an
intention-to-treat basis, that is we attempted to include all participants randomised to each group in the analyses, and all participants were analysed in the group to which they were allocated regardless of whether or not they received the allocated intervention.
The denominator for each outcome in each trial was the number
randomised minus any participants whose outcomes were known
to be missing.
Measures of treatment effect
Assessment of heterogeneity
Dichotomous data
For dichotomous data, we presented the results as summary risk
ratio (RR) with 95% confidence interval (CI).
Continuous data
For continuous data, we used the mean difference for outcomes
measured in the same way between trials. We used the standardised
mean difference to combine trials that measured the same outcome
but used different scales.
We assessed statistical heterogeneity in each meta-analysis using
the Tau² (tau-squared), I², and X² (Chi²) statistics. We regarded
heterogeneity as substantial if:
(a) the I² value was high (exceeding 30%); and
either
(b) there was inconsistency between trials in direction or magnitude of effects (judged visually), or a low (< 0.10) P value in the
Chi² test for heterogeneity;
or
(c) the estimate of between-study heterogeneity ( Tau² ) was above
zero.
Assessment of reporting biases
Unit of analysis issues
Cluster-randomised trials
We did not identify any cluster-randomised trials in our search. In
future updates of this review, if we identify any cluster-randomised
trials we will include them along with individually randomised trials. We will adjust their sample sizes using the methods described
in the Handbook for Systematic Reviews of Interventions (Higgins
2011) using an estimate of the intracluster correlation co-efficient
(ICC) derived from the trial (if possible), from a similar trial, or
from a study of a similar population. If we use ICCs from other
sources we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we identify cluster-
As there were less than 10 studies included in the meta-analysis we
did not investigate publication bias using funnel plots. In future
updates of this review, if there are 10 or more studies included in
the meta-analysis, we will assess funnel plot asymmetry visually
and use formal tests for funnel plot asymmetry. For continuous
outcomes, we will use the test proposed by Egger 1997, and for
dichotomous outcomes we will use the test proposed by Harbord
2006. If asymmetry is detected in any of these tests or is suggested
by a visual assessment, we will performed exploratory analyses to
investigate it.
Data synthesis
We carried out statistical analysis using the Review Manager software (RevMan 2011). As there was clinical diversity in respect of
7
the interventions, that is each trial used a different topical preparation with different active ingredients, we used a random-effects
model meta-analysis to produce an overall summary of the average
treatment effect across the six included trials. This random-effects
summary is treated as the average range of possible treatment effects and therefore the true effect differs in the different trials or
varies ’...across studies about an overall pooled value’ (Riley 2011).
For each outcome reported, we presented the results of the random-effects model analyses as the average treatment effect with its
95% confidence interval, and the estimates of Tau² and I².
Subgroup analysis and investigation of heterogeneity
We neither planned nor conducted any subgroup analyses.
In future updates of this review, if we identify substantial heterogeneity we will investigate it using subgroup analyses and sensitivity analyses. We will consider whether an overall summary is
meaningful and, if it is, use random-effects model analysis to produce it.
We plan to carry out the following subgroup analysis.
1. Parity (nulliparous versus multiparous women).
Subgroup analysis will be restricted to primary outcomes.
We will assess subgroup differences by the interaction tests available within RevMan (RevMan 2011). We will report the results
of subgroup analyses quoting the χ 2 statistic and P value, and the
interaction statistic I² value.
Results of the search
Our updated search identified a total of 13 reports relating to 12
trials (Buchanan 2010; de Buman 1987; Lachmann 2011; Mallol
1991; Martius 1973; Mendez Velarde 2010; Msika 2002; Ortega
1985; Osman 2008; Puder 1965; Taavoni 2011; Wierrani 1992).
Eight new potential trials for inclusion were identified (Buchanan
2010; de Buman 1987; Lachmann 2011; Mendez Velarde 2010;
Msika 2002; Ortega 1985; Osman 2008; Taavoni 2011) in addition to the four studies (Mallol 1991; Martius 1973; Puder
1965; Wierrani 1992) included in the previous version of this review (Young 1996). All of the trials were retrieved via the search
of the Cochrane Pregnancy and Childbirth Group Trials Register with the exception of two studies (Lachmann 2011; Msika
2002). Msika 2002 was found from searching reference lists of
retrieved studies and Lachmann 2011 via communication with
Expanscience Laboratories in France during our attempts to get
more information on Msika 2002. Other searches did not yield
any further potentially eligible studies.
This updated review includes six studies (involving 800 women).
Two additional trials (Lachmann 2011; Ortega 1985) are awaiting
classification. Despite translating the paper, there was insufficient
information on the randomisation process to judge if the study
by Ortega 1985 was eligible for inclusion and attempts to contact
the authors have been unsuccessful (see Studies awaiting classification). We are awaiting the translated report of Lachmann 2011
from Expanscience Laboratories (France).
Included studies
Sensitivity analysis
The previous version of this review (Young 1996) did not include
any a priori sensitivity analysis. In this update we undertook a
sensitivity analysis by trial quality, by removing from the analysis
those studies judged to be at high risk of bias for random sequence
generation, allocation concealment, or with more than 20% missing data for a given outcome. In future updates of this review,
the criteria for sensitivity analysis will broaden to determine the
effect of also excluding trials judged to be at high risk of bias for
blinding.
RESULTS
Description of studies
See: Characteristics of included studies; Characteristics of excluded
studies; Characteristics of studies awaiting classification.
This update includes four (Buchanan 2010; de Buman 1987;
Osman 2008; Taavoni 2011) new studies bringing the total number of included studies to six (involving a total of 800 women)
(Buchanan 2010; de Buman 1987; Mallol 1991; Osman 2008;
Taavoni 2011; Wierrani 1992) (see Characteristics of included
studies).
Included studies were undertaken in Germany (de Buman 1987),
Spain (Mallol 1991), Austria (Wierrani 1992), West Indies
(Buchanan 2010), Lebanon (Osman 2008) and Iran (Taavoni
2011) and were conducted mainly in antenatal clinics and medical
centres.
Two studies compared topical preparations with active ingredients with placebo, that is Trofolastin (which contains Centella asiatica extract, alpha tocopherol and collagen-elastin hydrolysates)
versus placebo cream (Mallol 1991) and cocoa butter lotion versus placebo lotion (Osman 2008). Two studies compared topical preparations with active ingredients with no treatment, that is
olive oil versus no treatment (Taavoni 2011) and Verum ointment
(which contains vitamin E, essential free fatty acids (vitamin F),
panthenol, hyaluronic acid, elastin and menthol) versus no treatment (Wierrani 1992). One study compared topical preparations
with active ingredients with other topical preparations with active ingredients, that is cocoa butter cream versus a similar cream
8
with vitamin E and other constituents but without cocoa butter
(Buchanan 2010). Finally, one study included two intervention
groups and a placebo group; topical preparations with active ingredients were compared with other topical preparations with active ingredients, and topical preparations with active ingredients
were compared with placebo, that is Alphastria cream (which contains hyaluronic acid, vitamin A, vitamin E, allantoine, calcium
pantothenate) versus a cream with vitamins and excipients, and
Alphastria cream and cream with vitamins and excipients versus a
cream with excipients only (de Buman 1987).
Excluded studies
For this update, we have added two new excluded studies (Mendez
Velarde 2010; Msika 2002) bringing the total number of excluded
studies to four (Martius 1973; Mendez Velarde 2010; Msika 2002;
Puder 1965) (see Characteristics of excluded studies).
Risk of bias in included studies
We assessed the risk of bias in the included studies (Figure 1; Figure
2). Overall, the studies were at low or unclear risk of bias across
most domains (Figure 1). No study was at low risk of bias across
all seven domains, while three studies (Buchanan 2010; Mallol
1991; Osman 2008) were at low risk of bias across five of the
seven domains and one study (Taavoni 2011) was at low risk of
bias across four of the seven domains. All of the included studies
were at unclear risk of bias for allocation concealment except for
Wierrani 1992, which was at high risk of bias (Figure 2). Details
of risk of bias within domains and across studies are given below.
Figure 1. Risk of bias graph: review authors’ judgements about each risk of bias item presented as
percentages across all included studies.
9
Figure 2. Risk of bias summary: review authors’ judgements about each risk of bias item for each included
study.
10
Allocation
Four of the six included studies (Buchanan 2010; Mallol 1991;
Osman 2008; Taavoni 2011) were at low risk of bias in random
sequence generation, one was at unclear risk (de Buman 1987)
and one was at high risk (Wierrani 1992). de Buman 1987 was
at unclear risk due to insufficient information to judge the risk of
bias, while Wierrani 1992 was at high risk of bias as alternate day
allocation was used to enrol women in the treatment groups. On
uneven dates women were included in the treatment group and
on even dates women were enrolled in the no treatment group.
Five of the six included studies (Buchanan 2010; de Buman 1987;
Mallol 1991; Osman 2008; Taavoni 2011) were at unclear risk of
bias in allocation concealment and Wierrani 1992 was at high risk
of bias. Two of the six studies, with unclear risk of bias for allocation
concealment (Buchanan 2010; de Buman 1987), had insufficient
information to judge the risk of bias while three of the other four
(Mallol 1991; Osman 2008; Taavoni 2011) reported insufficient
information on how women were allocated to the study groups.
In Wierrani 1992 randomisation was performed according to day
of treatment (that is, alternate days).
Blinding
Three of the six included studies (Buchanan 2010; Mallol 1991;
Osman 2008) were at low risk of bias in blinding of participants
and personnel and in blinding of outcome assessment. Three of
the six included studies were at unclear risk (de Buman 1987;
Taavoni 2011; Wierrani 1992) due to insufficient or no information reported to permit judgement of risk of bias for performance
or detection bias.
Incomplete outcome data
Five of the six included studies (Buchanan 2010; de Buman 1987;
Mallol 1991; Osman 2008; Taavoni 2011) were at low risk of
bias for incomplete outcome data and one study was at unclear
risk (Wierrani 1992). In the study by Wierrani 1992 the number of women randomised was not given, only the number of
women included in the analysis. In the Mallol 1991 trial report
it was not stated explicitly how many women were randomised,
but it was stated that ‘The assay was carried out on 100 pregnant women’. They reported total ‘valid’ cases as 41 for intervention (active cream) and 39 for placebo. Assuming 100 women
were randomised, then 20 women were excluded from the analysis overall, giving 20% incomplete outcome data (low risk) (see
Characteristics of included studies).
Selective reporting
All of the six included studies (Buchanan 2010; de Buman 1987;
Mallol 1991; Osman 2008; Taavoni 2011; Wierrani 1992) were
at low risk of bias in selective reporting.
Other potential sources of bias
We judged two of the six included studies (de Buman 1987;
Taavoni 2011) to be at low risk of other biases, and four studies
(Buchanan 2010; Mallol 1991; Osman 2008; Wierrani 1992) at
unclear risk of other biases. In the studies by Mallol 1991 and
Wierrani 1992, the number of women randomised to each group
were not stated explicitly, while in the studies by Buchanan 2010;
and Osman 2008 the researchers raised concerns regarding intervention fidelity. Buchanan 2010: 68 stated ’...that it was not possible to verify that the patients were using the cream as instructed
or whether they were sharing the cream or using other creams’,
while Osman 2008: 1142 stated that ’Reports of compliance varied greatly for each patient and assessors reported that women may
have been telling them what they wanted to hear when they asked
about compliance’. They acknowledged that compliance may have
been an issue in the study but that use of the study lotion reflected general population use (Osman 2008: 1142). Therefore,
we judged Buchanan 2010 and Osman 2008 to be at unclear risk
of other biases.
The source of funding was not identified for all but two of the included studies, Osman 2008 where both intervention and placebo
lotions were provided by ET Browne Drug Company, Inc and
Taavoni 2011 where the authors declared no funding source.
Effects of interventions
This updated review now includes data from six studies involving
800 women.
1. Topical preparations with active ingredients
compared with placebo or no treatment (five trials
and 474 women)
This comparison includes data from the following studies: de
Buman 1987; Mallol 1991; Osman 2008; Taavoni 2011; Wierrani
1992.
Primary outcome (five trials and 474 women)
There was no statistically significant average difference in the development of stretch marks in women who received topical preparations with active ingredients compared to women who received a
placebo or no treatment (average risk ratio (RR) 0.74; 95% confidence interval (CI) 0.53 to 1.03; five trials, 474 women; randomeffects model, Tau² = 0.09, I² = 65%) (Analysis 1.1).
11
Results were consistent with the main effects when we performed a
sensitivity analysis, excluding the Wierrani 1992 study which was
at high risk of bias for random sequence generation and allocation
concealment (average RR 0.81; 95% CI 0.60 to 1.10; four trials,
424 women; random-effects model, Tau² = 0.05, I² = 57%).
Secondary outcome (two trials and 255 women)
There was no statistically significant average mean difference in
the severity of stretch marks in women who received topical preparations with active ingredients compared to women who received
a placebo or no treatment (standardised mean difference (SMD)
-0.31; 95% CI -1.06 to 0.44; two trials, 255 women; randomeffects model, Tau² = 0.26, I² = 87%) (Analysis 1.2). The heterogeneity between the two trials was large and therefore the average
result may not be meaningful, that is it is unlikely that such an
average effect would be found in real life (the effect could be similar to one or other of the trials, but unlikely to be an ’average’ of
both).
2. Topical preparations with active ingredients
compared with other topical preparations with active
ingredients (two trials and 305 women)
This comparison included two studies with 305 women (
Buchanan 2010; de Buman 1987). Buchanan 2010 compared cocoa butter cream, which contained a variety of constituents for
example 25% cocoa butter, glycerin, isopropyl palmitate, hydrolysed collagen, hydrolysed elastin and tocopheryl acetate (vitamin
E) with a cream identical to the intervention cream but without
the 25% cocoa butter. In the second trial (de Buman 1987), Alphastria cream (containing hyaluronic acid, vitamin A, vitamin
E, allantoine, calcium pantothenate) was compared with a cream
with vitamins and excipients.
Primary outcome
There was no statistically significant average difference in the development of stretch marks in women who received topical preparations with active ingredients compared to women who received
other topical preparations with active ingredients (average RR
0.51; 95% CI 0.16 to 1.60; two trials, 305 women; random-effects model, Tau² = 0.53, I² = 74%) (Analysis 2.1).
Secondary outcome (one trial and 206 women)
The was no statistically significant mean difference in the severity
of stretch marks in women who received topical preparations with
active ingredients compared to women who received other topical
preparations with active ingredients (mean difference (MD) -0.20;
95% CI -0.53 to 0.13; one trial, 206 women; heterogeneity not
applicable) (Analysis 2.2).
DISCUSSION
Stretch marks are a very common connective tissue change that
can occur in pregnancy (Lawley 1999), affecting between 50% and
90% (Osman 2007) of women and remaining as silvery scar lines
on the skin. They are an unwanted consequence of pregnancy,
impacting on women’s perception of themselves (Osman 2008)
and their quality of life (Salter 2006), and are thus of significant
concern to women of child bearing age. There are many products
of unproven effectiveness on the market (Burrows 2010), which
are tried by many women. This review assessed the effects of topical
preparations on the prevention of stretch marks in pregnancy.
The review includes six trials (involving a total of 800 women)
conducted mainly in antenatal clinics and medical centres in varied geographical locations. It should be noted that the randomeffects model summaries presented are the average effects found
for both the development and severity of stretch marks. The use
of the random-effects model is based on the assumption that the
treatment effect will be different across the studies due to heterogeneity between the studies. It therefore calculates the average of
all the treatment effects across the trials (Riley 2011) and thus may
not be the actual effect in any of the included studies.
Summary of main results
Topical preparations with active ingredients
compared with placebo or no treatment
This review found that there was no statistically significant average difference in the development of stretch marks in women
who received topical preparations with active ingredients compared to women who received a placebo or no treatment. All studies were relatively small, with four of the five trials including less
than 100 women (de Buman 1987; Mallol 1991; Taavoni 2011;
Wierrani 1992) and one trial by Osman 2008 including less than
200 women as reflected in the narrower confidence interval.
Trials differed in relation to the timing of commencement of the
topical applications. Three of the five studies included women
presenting in trimester one (de Buman 1987; Mallol 1991; Osman
2008) while Taavoni 2011 and Wierrani 1992 recruited women
at 18 to 20 weeks and 20 weeks’ gestation, respectively. All trials
recruited women prior to the third trimester when stretch marks
usually occur (Atwal 2006; Cunningham 2010).
Parity of the women participating in the included trials also differed. Mallol 1991 included both multigravidae and primigravidae, while Taavoni 2011 and Osman 2008 included primigravidae women only in their studies. The study by de Buman 1987
identifies that one case of stretch marks occurred in a twin pregnancy in the intervention group receiving with active ingredients
(Group A). No information on parity was given for the other included study (Wierrani 1992). It is likely, therefore, that some of
12
the women in some trials had stretch marks from an earlier pregnancy, while the inclusion of some women with a multiple pregnancy may have increased the likelihood of those women developing stretch marks. In a multiple pregnancy there may be extra
strain and, as suggested by Watson 1998, extra strain or stretching
could be sufficient to tear the elastic fibre network, resulting in the
formation of striae.
While topical preparations varied, many of them included some
common ingredients. For example, two trials (de Buman 1987;
Wierrani 1992) included hyaluronic acid in the active topical
preparation (that is Alphastria cream and Verum ointment, respectively). Hyaluronic acid has been identified as stimulating fibroblastic activity (de Buman 1987), and fibroblasts are key cells
in maintaining tissue structure and tone. A number of the preparations also contained vitamin E (de Buman 1987; Mallol 1991;
Osman 2008; Taavoni 2011; Wierrani 1992), which promotes the
development of the intracellular substance (Wierrani 1992) and is
a known antioxidant used in many skin products. However, it is
not evident which, if any, of these ingredients could exert a possible preventative action for the formation of striae.
We found no statistically significant average difference in the
severity of stretch marks in women who received topical preparations with active ingredients compared to women who received a
placebo or no treatment. Data on the severity of stretch marks were
only available in two of the five included studies (Mallol 1991;
Osman 2008). In the study by Osman 2008, severity was assessed
by trained assessors using a validated tool, while Mallol 1991 refers
to using ’an arbitrary score 0 = no striae, 1 = few and thin striae, 2
= many thin striae or few thick striae, and 3 = many thick striae’.
Neither study refers to inter-rater reliability and therefore it is unclear how errors in measurement were minimised.
From a clinical perspective while none of the topical products in
the included trials (Alphastria cream, Trofolastin, cocoa butter lotion, olive oil, and Verum), with the exception of cocoa butter
lotion (Osman 2008) and olive oil (Taavoni 2011), appear to be
widely available, some women may indeed be applying cocoa butter lotion or olive oil in the hope of preventing the development
of stretch marks. However, this review found no statistically significant average difference in the development of stretch marks
in women who received topical preparations with active ingredients compared to women who received a placebo or no treatment.
Therefore, based on this review it is not possible to recommend
any of the preparations.
Topical preparations with active ingredients
compared with other topical preparations with active
ingredients
This review found no statistically significant average difference
in the development of stretch marks or in the severity of stretch
marks in women who received topical preparations with active
ingredients compared with other topical preparations with active
ingredients in trials (Buchanan 2010; de Buman 1987) involving
small numbers of participants. In the trial by Buchanan 2010, the
topical preparations contained multiple ingredients including isopropyl palmitate (emollient), propylene glycol isostearate (emollient), PPG-15 stearyl ether (1-octadecoxypropan-2-ol) (emollient), hydrolysed collagen, hydrolysed elastin and tocopheryl acetate (vitamin E). Preparations differed only in the addition of cocoa butter (25%). In the study by de Buman 1987, the intervention preparation contained several ingredients (hyaluronic acid,
vitamin A, vitamin E, allantoine, calcium pantothenate) while the
comparison preparation contained vitamins in addition to the excipient.
Data on the severity of stretch marks were only available in one
of the two included studies (Buchanan 2010). In this study,
Buchanan 2010 assessed the severity of stretch marks using the
’4 quadrant technique of Davey (1972) with a simplification by
Fletcher (unpublished)’. The researchers outline the process of ensuring that assessments were reliable and state that checking of
researchers’ assessments was undertaken with the aid of digital
photographs until such time as ’a consensus of the striae scoring
system by different observers’ was achieved.
Overall completeness and applicability of
evidence
Based on the findings of this review, which included six small trials,
it is not possible to recommend any of the preparations for the
prevention of stretch marks in pregnancy .
Quality of the evidence
We assessed the risk of bias of included trials as ’low risk’ for
random sequence generation, blinding of participants and personnel, and in blinding of outcome assessment, complete outcome data and selective reporting in only three of the six trials
(Buchanan 2010; Mallol 1991; Osman 2008). We assessed one
study (Wierrani 1992) as at high risk of bias for random sequence
generation and allocation concealment. Overall findings are not
sensitive to exclusion of this study.
The quality of evidence is also impacted on by the possible imprecision of the study results due to the small numbers of participants
and events, and their wide confidence intervals. This is particularly
evident in some of the trials (Taavoni 2011; Wierrani 1992).
Potential biases in the review process
We have taken every step to ensure that there are no potential biases
in the review processes. We undertook a systematic and comprehensive search without language restrictions and adhered to best
practice in undertaking the review. Three authors (MB, DD and
GY) independently assessed each study and agreed the studies that
13
were for inclusion in the review and those that were for exclusion.
Data extraction was also completed independently and checked
for accuracy by three authors (MB, DD and GY). We contacted
authors from two trials (Horace Fletcher for Buchanan 2010;
P Msika for Msika 2002) for further information (see notes in
Characteristics of included studies and Characteristics of excluded
studies). All data were entered into the Review Manager software
(RevMan 2011) and checked for accuracy by three authors (MB,
DD and GG). Consequently, biases in the review processes are
unlikely.
Preparations possibly worth pursuing might include Trofolastin
(Mallol 1991), Alphastria (de Buman 1987), and Verum (Wierrani
1992). The latter two preparations contain hyaluronic acid, which
has been identified as stimulating fibroblastic activity (de Buman
1987) and therefore maintaining tissue structure and tone. In addition, it is important that preparations commonly used by women
to prevent and treat stretch marks are evaluated within the context of robust, methodologically rigorous and adequately powered
randomised trials.
AUTHORS’ CONCLUSIONS
ACKNOWLEDGEMENTS
Implications for practice
We found no high-quality evidence to support the use of any of
the topical preparations in the prevention of stretch marks during
pregnancy.
Implications for research
There is a clear need for robust, methodologically rigorous randomised trials involving larger sample sizes to evaluate the effects
of topical preparations on the development of stretch marks in
pregnancy.
We would like to acknowledge the assistance of Luciana Figuera,
Charlese Allen, Caroline Summers and Gloria Avalos who kindly
did some translations of non-English language papers for us.
We thank David Jewell for his contribution to the original version
of this review.
As part of the pre-publication editorial process, this review has
been commented on by four peers (an editor and three referees
who are external to the editorial team), a member of the Pregnancy
and Childbirth Group’s international panel of consumers and the
Group’s Statistical Adviser.
REFERENCES
References to studies included in this review
Buchanan 2010 {published data only}
Buchanan K, Fletcher HM, Reid M. Prevention of striae
gravidarum with cocoa butter cream. International Journal
of Gynecology and Obstetrics 2010;108(1):65–8.
de Buman 1987 {published data only}
de Buman M, Walther M, de Weck R. Effectiveness of
Alphastria cream in the prevention of pregnancy stretch
marks (striae distensae). Results of a double-blind study
[Wirksamkeit der Alphastria–Creme bei der Vorbeugung
von Schwangerschaftsstreifen (Striae distensae). Ergebnisse
einer Doppelblind–Studie]. Gynakologische Rundschau
1987;27:79–84.
Mallol 1991 {published data only}
Belda MA, Costa D, Noval A, Sola M, Mallol J. Prophylaxis
of striae gravidarum with a topical formulation - a double
blind study. Proceedings of 22nd International Congress
of Confederation of Midwives; 1990 October 7-12; Kobe,
Japan. 1991.
Mallol J, Belda MA, Costa D, Noval A, Sola M. Prophylaxis
of striae gravidarum with a topical formulation. A double
blind trial. International Journal of Cosmetic Science 1991;3:
51–7.
Osman 2008 {published data only}
Osman H. Cocoa butter for prevention of stretch marks.
ClinicalTrials.gov (http://clinicaltrials.gov/ct2/show/
NCT00114660) (accessed 21 March 2006).
Osman H, Usta IM, Rubeiz N, Abu-Rustum R, Charara
I, Nassar AH. Cocoa butter lotion for prevention of striae
gravidarum: a double-blind, randomised and placebocontrolled trial. BJOG: an international journal of obstetrics
and gynaecology 2008;115(9):1138–42.
Taavoni 2011 {published data only}
Taavoni S, Soltanipour F, Haghani H, Ansarian H,
Kheirkhah M. Effects of olive oil on striae gravidarum in
the second trimester of pregnancy. Complementary Therapies
in Clinical Practice 2011;17(3):167–9.
Wierrani 1992 {published data only}
Wierrani F, Kozak W, Schramm W, Grunberger W.
Attempt of preventive treatment of striae gravidarum
using preventive massage ointment administration. Wiener
Klinische Wochenschrift 1992;104:42–4.
References to studies excluded from this review
14
Martius 1973 {published data only}
Martius G. Prevention of striae gravidarum. Medizinische
Welt 1973;24:799–800.
Mendez Velarde 2010 {published data only}
Méndez Velarde FA, Salceda Pérez MA, Moya CB, Azpeitia
De La O CM, Vázquez Coronado DP, Ore Colio L.
Effect of a moisturizing cream in the prevention of striae
gravidarum [Efecto de una crema hidratante para prevenir
las estrias del embarazo]. Dermatologia Revista Mexicana
2010;54(5):273–8.
Msika 2002 {published data only}
Msika P, Gavaud-Kennedy C, Pitiot M, Camel E, ArnaudBoissel L, Guillot JP. Prevention of stretch marks: A
randomized double bind clinical study versus placebo.
Journal of European Academy of Dermatology and Venereology
2002;16 Suppl 1:154.
Msika P, Gavaud-Kennedy C, Pitiot M, Camel E, ArnaudBoissel L, Guillot JP. Prevention of stretch marks: a
randomized double blind clinical study versus placebo.
http://www.mustelausa.com/professional/˙images/pdfs/
M9.pdf (accessed 5 October 2011) 2002.
Puder 1965 {published data only}
Puder H. Treatment of striae gravidarum. Medizinische Welt
1965;16:650–3.
References to studies awaiting assessment
Lachmann 2011 {published data only}
Lachmann N, Biassette M, Boy E, Chadoutaud B, Msika P.
Assessment of the preventive effect of a topical treatment
on striae gravidarum during pregnancy: a multicenter open
study on primiparae. Journal of Investigative Dermatology
September 2011; Vol. 131, issue Issue S2, Abstract 275:
S46.
Ortega 1985 {published data only}
Ortega A. Clinical and experimental study of the prevention
of stretch marks in pregnancy [Estudio clínico experimental
de la prevención de las estrías en el embarazo]. Acta
Ginecologica 1985;42(1):29–36.
Additional references
Alster 1997
Alster TS. Laser treatment of hypertrophic scars, keloids,
and striae. Dermatologic Clinics 1997;15(3):419–29.
Archer 2004
Archer CB. Functions of the skin. In: Burns T, Breathnach
S, Cox N, Griffiths C editor(s). Rook’s Textbook of
Dermatology. 7th Edition. Vol. 1, Malden Massachusetts:
Blackwell Science, 2004:4.1–4.12.
Atwal 2006
Atwal GSS, Manku LK, Griffiths CEM, Polson DW. Striae
gravidarum in primiparae. British Association of Dermatology
2006;55(5):965–9.
Brinkhaus 2000
Brinkhaus B, Lindner M, Schuppan D, Hahn EG.
Chemical, pharmacological and clinical profile of the East
Asian medical plant Centella asiatica. Phytomedicine 2000;7
(5):427–48.
Bryant 1968
Bryant WM, Greenwell JE, Weeks PM. Alterations in
collagen organization during dilatation of the cervix uteri.
Surgery, Gynecology & Obstetrics 1968;126(1):27–39.
Burrows 2004
Burrows NP, Lovell CR. Disorders of connective tissue. In:
Burns T, Breathnach S, Cox N, Griffiths C editor(s). Rook’s
Textbook of Dermatology. 7th Edition. Vol. 3, Malden
Massachusetts: Blackwell Science, 2004:46.1–46.71.
Burrows 2010
Burrows NP, Lovell CR. Disorders of connective tissue. In:
Burns T, Breathnach S, Cox N, Griffiths C editor(s). Rook’s
Textbook of Dermatology. 8th Edition. Vol. 3, Malden
Massachusetts: Blackwell Science, 2010:45.1–45.70.
Chang 2004
Chang ALS, Agredano YZ, Kimball AB. Risk factors
associated with striae gravidarum. Journal of the American
Academy of Dermatology 2004;51(6):881–5.
Clarke 2000
Clarke M, Oxman AD, editors. Cochrane Reviewers’
Handbook 4.2 [updated February 2003]. In: The
Cochrane Library [database on CDROM]. The Cochrane
Collaboration. Oxford: Update Software; 2003, Issue 2.
Cunningham 2010
Cunningham FG, Leveno KJ, Bloom SL, Hauth JC, Rouse
DJ, Spong CY. Williams Obstetrics. 23rd Edition. New
York: McGraw Hill, 2010.
Egger 1997
Egger M, Davey Smith G, Schneider M, Minder C. Bias
in meta-analysis detected by a simple, graphical test. BMJ
1997;315(7109):629–34.
Elsaie 2009
Elsaie ML, Baumann LS, Elsaaiee LT. Striae distensae
(stretch marks) and different modalities of therapy: an
update. Dermatologic Surgery 2009;35(4):563–73.
Elson 1990
Elson ML. Treatment of striae distensae with topical
tretinoin. Journal of Dermatological Surgical Oncology 1990;
16(3):267–70.
Errickson 1994
Errickson CV, Matus NR. Skin disorders of pregnancy.
American Family Physician 1994;49(3):605–10.
Ghasemi 2007
Ghasemi A, Gorouhi F, Rashighi-Firoozabadi M, Jafarian S,
Firooz A. Striae gravidarum: associated factors. Journal of
the European Academy of Dermatology and Venereology 2007;
21(6):743–6.
Harbord 2006
Harbord RM, Egger M, Sterne JA. A modified test for
small-study effects in meta-analyses of controlled trials
with binary endpoints. Statistics in Medicine 2006;25(20):
3443–57.
15
Higgins 2011
Higgins JPT, Green S, editors. Cochrane Handbook for
Systematic Reviews of Interventions Version 5.1.0 [updated
March 2011]. The Cochrane Collaboration, 2011.
Available from www.cochrane-handbook.org.
Horn 2007
Horn TD, Hoskyn J. Dermatologic disorders during
pregnancy. In: Reece A, Hobbins JC editor(s). Clinical
Obstetrics. The Fetus & Mother. 3rd Edition. Malden
Massachusetts: Blackwell Publishing, 2007:947–60.
Hunter 1973
Hunter JAA. Diseases of the skin. The basis of skin therapy.
British Medical Journal 1973;4(5889):411–3.
Kang 1996
Kang S, Kim KJ, Griffith CEM, Wong TY, Talwar HS,
Fisher GJ, et al.Topical Tretinoin (retinoic acid) improves
early stretch marks. Archives of Dermatology 1996;132(5):
519–26.
Kang 1998
Kang S. Topical tretinoin therapy for management of early
striae. Journal of the American Academy of Dermatology
1998;39(2 Pt 3):s90–s92.
Lawley 1999
Lawley TJ, Yancey KB. Skin changes and diseases in
pregnancy. In: Freedberg IM, Eisen AZ, Wolf K, Austen
KF, Goldsmith LA, Katz SI, et al. editor(s). Fitzpatrick’s
Dermatology in General Medicine. 5th Edition. Vol. Vol II,
New York: McGraw Hill, 1999:1963–9.
Liu 1974
Liu DTY. Letter: Striae gravidarum. Lancet 1974;1(7858):
625.
Papoutsis 2007
Papoutsis J, Kroumpouzos G. Dermatologic disorders of
pregnancy. In: Gabbe SG, Niebyl JR, Simpson JL editor
(s). Obstetrics. Normal and Problem Pregnancies. 5th
Edition. Philadelphia: Churchill Livingstone Elsevier,
2007:1178–92.
Poidevin 1959
Poidevin LOS. Striae gravidarum. Their relation to adrenal
cortical hyperfunction. Lancet 1959;2(7100):436–9.
Rangel 2001
Rangel O, Arias I, Garcia E, Lopez-Padilla S. Topical
tretinoin 0.1% for pregnancy-related abdominal striae: An
open-label, multicenter, prospective study. Advances in
Therapy 2001;18(4):181–6.
RevMan 2011
The Nordic Cochrane Centre, The Cochrane Collaboration.
Review Manager (RevMan). 5.1. Copenhagen: The Nordic
Cochrane Centre, The Cochrane Collaboration, 2011.
Riley 2011
Riley D, Gates S, Neilson J, Alfirevic Z. Statistical methods
can be improved within Cochrane pregnancy and childbirth
reviews. Journal of Clinical Epidemiology 2011;64(6):
608–18.
Salter 2006
Salter SA, Kimball AB. Striae gravidarum. Clinics in
Dermatology 2006;24(2):97–100.
Shuster 1979
Shuster S. The cause of striae distensae. Acta DermatoVenereologica. Supplementum 1979;59(85):161–9.
McKenzie 1971
McKenzie AW. Skin disorders in pregnancy. The Practitioner
1971;206(236):773–80.
Summers 2009
Summers B, Lategan M. The effect of a topically-applied
cosmetic oil formulation on striae distensae. South African
Family Practice 2009;51(4):332–6.
Murphy 1992
Murphy KW, Dunphy B, O’Herlihy C. Increased maternal
age protects against striae gravidarum. Journal of Obstetrics
and Gynaecology 1992;12(5):297–300.
Thomas 2004
Thomas RGR, Liston WA. Clinical associations of striae
gravidarum. Journal of Obstetrics and Gynaecology 2004;24
(3):270–1.
Murray 2009
Murray I, Hassall J. Change and adaptation in pregnancy.
In: Fraser DM, Cooper MA editor(s). Myles Textbook for
Midwives. 15th Edition. Edinburgh: Churchill Livingstone
Elsevier, 2009:189–225.
Venning 1946
Venning EH. Adrenal function in pregnancy. Endocrinology
1946;39:203–20.
Muzaffar 1998
Muzaffar F, Hussain I, Haroon TS. Physiologic skin changes
during pregnancy: a study of 140 cases. International
Journal of Dermatology 1998;37(6):429–31.
Watson 1998
Watson REB, Parry EJ, Humphries JD, Jones CJP, Polson
DP, Kielty CM, et al.Fibrillin microfibrils are reduced in skin
exhibiting striae distensae. British Journal of Dermatology
1998;138(6):931–7.
Nigam 1989
Nigam PK. Striae cutis distensae. International Journal of
Dermatology 1989;28(7):426–8.
Waugh 2010
Waugh A, Grant A. Ross and Wilson Anatomy and Physiology
in Health and Illness. 11th Edition. Edinburgh: Churchill
Livingstone, 2010.
Osman 2007
Osman H, Rubeiz N, Tamim H, Nassar AH. Risk factors
for the development of striae gravidarum. American Journal
of Obstetrics and Gynecology 2007;196(1):62.e1–62.e5.
Wong 1984
Wong RC, Ellis CN. Continuing medical education.
Physiologic skin changes in pregnancy. Journal of the
American Academy of Dermatology 1984;10(6):929–40.
16
Wong 1989
Wong RC, Ellis CN. Physiologic skin changes in pregnancy.
Seminars in Dermatology 1989;8(1):7–11.
Zheng 1985
Zheng P, Lavker RM, Kligman AM. Anatomy of striae.
British Journal of Dermatology 1985;112(2):185–93.
References to other published versions of this review
Young 1995
Young GL. A cream to prevent striae gravidarum. [revised
12 May 1993]. In: Enkin MW, Keirse MJNC, Renfrew MJ,
Neilson JP, Crowther C (eds.) Pregnancy and Childbirth
Module. In: The Cochrane Pregnancy and Childbirth
Database [database on disk and CDROM] The Cochrane
Collaboration; Issue 2, Oxford: Update Software, 1995.
Young 1996
Young G, Jewell D. Creams for preventing stretch marks in
pregnancy. Cochrane Database of Systematic Reviews 1996,
Issue 1. [DOI: 10.1002/14651858.CD000066]
∗
Indicates the major publication for the study
17
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Buchanan 2010
Methods
Study design: randomised controlled trial.
Duration of study: not stated.
Participants
Setting: antenatal clinic in the University Hospital, of the West Indies, Mona, Kingston,
Jamaica
Inclusion criteria: ’...primigravidas and multigravidas with no stretch marks’ enrolled
before 16 weeks’ gestation
Exclusion criteria: ’Women who were taking steroids and women with medical illnesses
that caused stretch marks...’; ’Women with a twin pregnancy or polyhydramnios...’
Participants randomised: 150 women were randomly assigned to the intervention
(cocoa butter group) and 150 women to the comparison group
Interventions
Experimental: application of 473 ml of cocoa butter cream containing cocoa butter
cream (25%), water, glycerin (skin conditioner), distearyldimonium chloride (skin conditioner), isopropyl palmitate (emollient), cetearyl alcohol (stabilizer), propylene glycol
isostearate (emollient), PPG-15 stearyl ether (1-octadecoxypropan-2-ol) emollient, hydrolysed collagen, hydrolysed elastin, tocopheryl acetate (vitamin E), dimethicone (skin
conditioner). Half a cap full of cream was applied to the 4 abdominal quadrants daily
(until used up)
Control: cream which was identical to the intervention cream with the exception of
addition of the 25% cocoa butter
We did not consider this a true placebo as it contains other active ingredients like vitamin
E
Outcomes
Outcomes considered in the review:
• presence of stretch marks;
• severity of stretch marks.
Notes
’...number of stretch marks was assessed using the 4 quadrant technique of Davey (1972)
with a simplification by Fletcher (unpublished), which involved using a pictorial chart
to aid the providers in using Davey’s technique. Digital photographs were taken of the
abdomen of some women...’
1 of the study authors Horace Fletcher confirmed that the percentage of women who
developed striae was erroneously included in the enrolment data and that none of the
woman had striae at enrolment
Risk of bias
Bias
Authors’ judgement
Random sequence generation (selection Low risk
bias)
Support for judgement
’The women were assigned cocoa butter
cream or placebo using a table of random
numbers...’
18
Buchanan 2010
(Continued)
Allocation concealment (selection bias)
Unclear risk
Insufficient information on which to judge
risk of bias.
Blinding of participants and personnel Low risk
(performance bias)
All outcomes
’The women and the researchers were
blinded to the allocation’
Blinding of outcome assessment (detection Low risk
bias)
All outcomes
’...The researchers were blinded to the allocation’.
Incomplete outcome data (attrition bias)
All outcomes
Low risk
Development of striae
In the intervention arm, 30 women were
excluded from analysis (lost to follow up
n = 28, discontinued intervention due to
rash n =2). The 2 women who discontinued treatment were returned to group denominator (n = 122). 18.7% incomplete
outcome data.
In the control arm 28 women were excluded from analysis (lost to follow up n =
27, discontinued intervention due to skin
rash n = 1). The 1 woman who discontinued due to skin rash (n = 1) was restored
to the group denominator (n = 123). 18%
incomplete outcome data.
Severity of striae
Intervention group: data available for 101
of the 120 women. Missing data = 15.8%
Control group: data available for 105 of the
122 women. Missing data = 13.9%
Selective reporting (reporting bias)
Low risk
The protocol was not available and following clarification from the authors all the
outcomes stated in the methods section
were reported adequately in the results.
Other bias
Unclear risk
Intervention integrity
Authors state ’...that it was not possible to
verify that the patients were using the cream
as instructed or whether they were sharing
the cream or using other creams’
Funding source
No funding source identified.
19
de Buman 1987
Methods
Duration of study: not stated (states women were monitored over 10 months).
Participants
Setting: Obstetrical and Gynecological Clinic in a Hospital in Cantonal, Fribourg
Inclusion criteria: women at the beginning of the third month of pregnancy.
Exclusion criteria: not stated.
Participants randomised: 90 women randomised: 30 women to each group (group A
- Alphastria cream n = 30; group B - cream with vitamins and excipients n = 30; group
C - placebo with just excipient n = 30)
Interventions
Experimental: ’Application of 10cm (3g) Alphastria cream (contains hyaluronic acid,
Vitamin A, Vitamin E, allantoine, calcium pantothenate)...’ daily to the thighs, abdomen
and chest. ’The cream was massaged gently into each area for a few minutes each’
Control: application of 1 of 2 creams : 1 containing vitamins and excipients, and 1
containing excipients only
Outcomes
• presence of stretch marks.
Notes
As group B contains vitamins and excipients we compared group A (Alphastria cream)
with group B (vitamins and excipients) and then group A and B with group C (excipients
only) [MB]
Risk of bias
Bias
Random sequence generation (selection Unclear risk
bias)
States that a randomised ’predetermined
code system’ was used. Insufficient Information on which to judge risk of bias.
Unclear risk
Treatments were administered anonymously. Insufficient information on which
to judge risk of bias
Blinding of participants and personnel Unclear risk
(performance bias)
All outcomes
While
treatments
were
administered anonymously and the study
is reported as a double blind study no detail
is given on who was blinded. Insufficient
detail on which to judge risk of bias.
Blinding of outcome assessment (detection Unclear risk
bias)
All outcomes
Insufficient information on which to judge
risk of bias.
All outcomes
In the control arm (group C) (n = 30)
, 3 women were excluded from the analysis (withdrawals due to intolerance, allergy, and miscarriage). We restored these 3
Low risk
20
de Buman 1987
(Continued)
women to the group denominator (n = 30)
.
Low risk
The protocol was not available but all the
were reported adequately in the results
Other bias
Low risk
None identified.
Funding source
No
funding
identified.
source
Mallol 1991
Methods
Duration of study: 30 months.
Participants
Setting: antenatal clinic in Barcelona.
Inclusion criteria: women in the first 12 weeks of pregnancy.
Exclusion criteria: none stated.
Participants randomised: not stated, study does not state how many were randomised
to the intervention and control group. Total valid cases are reported as 41 in the active
cream group and 39 in the placebo group
Interventions
Experimental: application of active cream (Trofolastin) (n = 41) which ’...was a marketed product...’ containing Centella asiatica extract and alpha- tocopherol and collagen - elastin hydrolysates. Application of active cream (Trofolastin) (n = 41) containing
Centella asiatica extract and alpha- tocopherol and collagen - elastin hydrolysates.
‘Product applied daily from the end of the 12th week of pregnancy to the day of labour
on abdomen, breasts, buttocks and hips’.
Control/Comparison intervention: placebo cream containing only the excipient part
of the active cream and ’...identical in colour, flavour and texture’.
‘Product applied daily from the end of the 12th week of pregnancy to the day of labour
on abdomen, breasts, buttocks and hips’
Outcomes
Notes
Striae were evaluated using ’...an arbitrary score 0 = no striae, 1 = few and thin striae, 2
= many thin striae or few thick striae, and 3 = many thick striae’
Risk of bias
Bias
bias)
Probably adequate as ‘randomised code
numbers’ were used.
21
Mallol 1991
(Continued)
Unclear risk
No details given as to how women were
allocated to the 2 groups
(performance bias)
All outcomes
Both intervention and control ’...creams
were identical in colour, flavour and texture’ and were ’...marked with a randomised code number’. Codes were not
opened until the data collection was complete
bias)
All outcomes
Codes were not opened until the data collection was complete
All outcomes
Low risk
States ‘The assay was carried out on 100
pregnant women’ but does not state explicitly how many women were randomised.
Total ‘valid’ cases are noted as 41 for intervention (active cream) and 39 for placebo.
Assuming 100 women randomised, then
20 women were excluded from the analysis
overall, due to abortion (n = 1) and ‘several
address changes’ (n = 19). 20% incomplete
outcome data.
Low risk
were reported in the results
Other bias
Unclear risk
Number of women randomised into each
group is not given. States ‘The assay was
carried out on 100 pregnant women’ but
does not state explicitly how many women
were randomised’
Funding source
Osman 2008
Methods
Duration of study: November 2004 to July 2006 (from study protocol).
Participants
Setting: 4 medical centres in Lebanon.
Inclusion criteria: nulliparous women with a singleton pregnancy presenting to the
clinic in trimester 1, ’...between November 2004 and December 2005
Exclusion criteria: women with a ’...known hypersensitivity to cocoa butter...’ or the
lotion components
Participants randomised: 105 women were randomised to the intervention group (co-
22
Osman 2008
(Continued)
coa butter lotion) and 105 to the placebo lotion group
Interventions
Experimental: application of a thin layer of a commercially available lotion containing
cocoa butter and tocopheryl acetate (vitamin E) to the abdomen, breasts and thighs once
daily, from ’...between 12 and 18 completed weeks of gestation...’ to delivery
Control: placebo lotion with no active ingredients, which ’...was made to look, smell,
and feel the same as the study lotion...’
Outcomes
Notes
Trained assessors (n = 5) completed ’... the data collection tools and...’ assessed ’...the
severity of SG based on a scale developed and previously validated by the authors (Osman
et al 2007). ’The scale took ...consideration the density and width of striae to estimate
the surface area of the body part affected’
’...funding by the Center for Research on Population and Health at the American University of Beirut, Lebanon, with generous support from the Wellcome Trust’ ’...and the
University Research Board (URB) at the American University of Beirut, Beirut, Lebanon’
Risk of bias
Bias
bias)
’Randomization was conducted using a
computer-generated random number table’
Not stated how participants were allocated
to each of the 2 groups
Unclear risk
(performance bias)
All outcomes
’...researchers, study participants and their
physicians were blinded to the lotion assignment’. ’...Codes for the study and
placebo lotions were opened after the final
assessment of the last randomised woman’
bias)
All outcomes
’Assessors... were blinded as to the lotion
assignment’. ’...Codes for the study and
placebo lotions were opened after the final
assessment of the last randomised woman’
All outcomes
In the intervention arm (n = 105), 14
women were excluded from analysis (remaining n = 91) due to abortion n = 3,
withdrew n = 2, lost to follow up n = 9.
We restored n = 3 (abortion) to the group
denominator (n = 94). Equates to 10.5%
incomplete outcome data in the interven-
Low risk
23
Osman 2008
(Continued)
tion group.
In the control arm (n = 105), 21 women
were excluded from analysis (remaining n
= 84) due to abortion n = 5, withdrew n =
6, lost to follow up n = 9, allergic reaction
n = 1). We restored n = 6 (abortion, and
allergic reaction) to the group denominator (n = 90). Equates to 14.3% incomplete
outcome data in the control arm
Low risk
Study protocol is available and all outcomes
have been reported as planned. However a
secondary outcome (severity of striae) is reported in the study which is not identified
as such in the protocol. Study protocol reports that women will be asked to give their
assessment of the presence or absence of
striae and their severity’. Trained assessors
assessed the severity of striae in the study.
Other bias
Unclear risk
Intervention integrity
Authors state that ’Reports of compliance
varied greatly for each patient and assessors
reported that women may have been telling
them what they wanted to hear when they
asked about compliance’
Funding source
Both study and placebo lotions were provided by ET Browne Drug Company, Inc
Taavoni 2011
Methods
Participants
Setting: Department of Obstetrics and Gynaecology, Lolagar Hospital and Shahid Akbarabadi Hospital, Iran University of Medical Sciences, Tehran, Iran
Inclusion criteria: ’...nulliparous women aged between 20 and 30 years old, in their
18th to 20th week of gestation with body mass indices ranging between BMI 18.5--25’.
Exclusion criteria: ’...included: [polyhydramnios], occurrence of dermal discuses, administration of corticosteroids, application of other ointments on the abdominal area,
lack of compliance with the study protocol’
Participants randomised: 35 women were randomised to the intervention group and
35 women to the control group
Interventions
Experimental: application of ’...olive oil topically onto...’ the ’... abdominal skin ...twice
a day...for eight weeks, ...without massaging’.
Application continued until week 28.
Control: no treatment.
24
Taavoni 2011
(Continued)
Outcomes
Outcomes considered in this review:
Notes
Risk of bias
Bias
bias)
’Subjects were randomised using a computer-generated randomization table to either the control or intervention group’
Unclear risk
No detail given as to how women were allocated to the 2 groups
(performance bias)
All outcomes
No information given on blinding in the
study.
bias)
All outcomes
No information given on blinding of outcome assessment.
All outcomes
Low risk
None (percentages in Table 2, p. 168 suggest no losses to follow up)
Low risk
The protocol was not available but the outcome stated in the methods section was reported adequately in the results
Other bias
Low risk
None identified.
Funding source
Authors declared no funding source.
Wierrani 1992
Methods
Study design: quasi-randomised controlled trial.
Participants
Setting: antenatal clinic in Vienna.
Inclusion criteria: pregnant women > 18 and < 35 years.
Exclusion criteria: history of metabolic disorders; long term medication use for example
corticosteroid use; alcohol abuse; history suggestive of a complicated pregnancy
Participants randomised: not stated. States that ’24 participants in the Verum group
and 26 in the control group could be included in the final evaluation’.
25
Wierrani 1992
(Continued)
Interventions
Experimental: application of an ointment (Verum, which contained: vitamin E, essential
free fatty acids (vitamin F), panthenol, hyaluronic acid, elastin and menthol. Frequency
of application not stated
Control: no treatment.
Outcomes
Notes
Risk of bias
Bias
Random sequence generation (selection High risk
bias)
Alternate day allocation. ’On days with uneven dates the pregnant women were included in the Verum (treatment) group’.
Women enrolled on even dates were given
no treatment
Allocation was performed according to day
of treatment (see above)
High risk
(performance bias)
All outcomes
Insufficient information to judge risk of
bias.
bias)
All outcomes
Insufficient information to judge risk of
bias.
All outcomes
Unclear risk
Number of women randomised not given.
States that ’24 participants in the Verum
group and 26 in the control group could
be included in the final evaluation’.
Low risk
were reported adequately in the results.
Other bias
Unclear risk
Number of women randomised into each
group is not given. States that ’24 participants in the Verum group and 26 in the
control group could be included in the final evaluation’
Funding source
26
Characteristics of excluded studies [ordered by study ID]
Study
Reason for exclusion
Martius 1973
Not stated whether this study was randomised. Review authors believe it was not and attempts to contact
the author have failed
Mendez Velarde 2010
The study does not fit the criteria for inclusion in the review. It compared the application of a topical cream
on wet skin versus its application on dry skin. Nor did it have a placebo or a no treatment group
Msika 2002
The study is at unclear risk of bias in five of the seven domains: random sequence generation, allocation
concealment, blinding of participants and personnel, blinding of outcome assessment and selective reporting.
While initial attempts to contact the authors were unsuccessful, eventual contact was made with Expanscience
Laboratories. However, it was not possible to get further information on the trial
Puder 1965
Not randomised.
Characteristics of studies awaiting assessment [ordered by study ID]
Lachmann 2011
Methods
Study design: unclear.
Participants
Setting: not stated.
Inclusion criteria: women between ’16-19 weeks of amenorrhea...’. ’Primipare were selected according [to] factors
which have been associated with the SG occurrence’
Participants randomised: unclear.
Interventions
Experimental: unclear.
Control: unclear.
Study involved the application of a cream which ’contains patented ingredients: lupeol, natural biopeptides and
arabinogalactane which counteract tissue inflammation and stimulate extracellular matrix (ECM) remodelling’
’The cream was applied twice daily during 5 months’.
Outcomes
Notes
Funded by Expanscience Laboratories.
27
Ortega 1985
Methods
Study design: unclear.
Participants
Setting: Obstetrics and Gynaecology Department, Mother-Infant Hospital, Palma de Mallorca
Inclusion criteria: women in the second trimester of pregnancy who were ’not obese (more than 10% overweight)’
Participants randomised: it is unclear if the women were randomised to the different groups
It states that 146 women were ’distributed into groups’: Group I (n = 61) was assigned to the cream with excipients
only (placebo), group II (n = 55) was assigned the cream with the active ingredients while group III (n = 30) was
assigned to the control arm [’neither instructed to massage nor use a cream’]
Interventions
Experimental: application of an ’L. anti-striae cream’, containing ’fitelenos (simulating factors of neo-elastogenesis
and transcutaneous penetrating factors)’ ...’once or twice a day’ ...’on the abdomen, legs and breasts, in a down-up
direction following the skin’s traction lines’ from the ’second trimester’ until the puerperium. ’They were advised to
undergo a massage once or twice a day each lasting from 5 to 10 minutes’. Participation in exercise is also referred to
but no details are given.
Control: application of a cream containing excipients only ’once or twice a day’, ’on the abdomen, legs and breasts,
in a down-up direction following the skin’s traction lines’ from the ’second trimester’ until the puerperium. ’They
were advised to undergo a massage once or twice a day each lasting from 5 to 10 minutes’. Participation in exercise
is also referred to but no details are given
Or no treatment [no massage or cream].
Outcomes
Notes
28
DATA AND ANALYSES
Comparison 1. Topical preparations with active ingredients compared with placebo or no treatment
Outcome or subgroup title
No. of
studies
No. of
participants
Statistical method
1 Presence of stretch marks
2 Severity of stretch marks
5
2
474
255
Risk Ratio (M-H, Random, 95% CI)
Std. Mean Difference (IV, Random, 95% CI)
Effect size
0.74 [0.53, 1.03]
-0.31 [-1.06, 0.44]
Comparison 2. Topical preparations with active ingredients compared with other topical preparations with active
ingredient
Outcome or subgroup title
No. of
studies
No. of
participants
1 Presence of stretch marks
2 Severity of stretch marks
2
1
305
206
Statistical method
Risk Ratio (M-H, Random, 95% CI)
Mean Difference (IV, Random, 95% CI)
Effect size
0.51 [0.16, 1.60]
-0.20 [-0.53, 0.13]
29
Analysis 1.1. Comparison 1 Topical preparations with active ingredients compared with placebo or no
treatment, Outcome 1 Presence of stretch marks.
Review:
Comparison: 1 Topical preparations with active ingredients compared with placebo or no treatment
Outcome: 1 Presence of stretch marks
Placebo or
no
treatment
Study or subgroup
Active preparations
n/N
n/N
de Buman 1987
15/60
10/30
14.4 %
0.75 [ 0.38, 1.46 ]
Mallol 1991
14/41
22/39
19.0 %
0.61 [ 0.36, 1.00 ]
Osman 2008
75/94
71/90
31.6 %
1.01 [ 0.87, 1.17 ]
Taavoni 2011
16/35
22/35
21.2 %
0.73 [ 0.47, 1.13 ]
Wierrani 1992
7/24
16/26
13.8 %
0.47 [ 0.24, 0.95 ]
Total (95% CI)
254
220
100.0 %
0.74 [ 0.53, 1.03 ]
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
Total events: 127 (Active preparations), 141 (Placebo or no treatment)
Heterogeneity: Tau2 = 0.09; Chi2 = 11.59, df = 4 (P = 0.02); I2 =65%
Test for overall effect: Z = 1.78 (P = 0.075)
Test for subgroup differences: Not applicable
0.05
0.2
1
Favours active prep.
5
20
Favours placebo
30
Analysis 1.2. Comparison 1 Topical preparations with active ingredients compared with placebo or no
treatment, Outcome 2 Severity of stretch marks.
Review:
Comparison: 1 Topical preparations with active ingredients compared with placebo or no treatment
Outcome: 2 Severity of stretch marks
Study or subgroup
Placebo or
no
treatment
Active preparations
Std.
Mean
Difference
Weight
N
Mean(SD)
N
Mean(SD)
Mallol 1991
41
1.42 (0.5)
39
2.13 (1.32)
47.4 %
-0.71 [ -1.16, -0.26 ]
Osman 2008
91
2.2 (1.8)
84
2.1 (1.8)
52.6 %
0.06 [ -0.24, 0.35 ]
Total (95% CI)
132
IV,Random,95% CI
Std.
Mean
Difference
IV,Random,95% CI
100.0 % -0.31 [ -1.06, 0.44 ]
123
-20
-10
0
10
20
Favours placebo
31
Analysis 2.1. Comparison 2 Topical preparations with active ingredients compared with other topical
preparations with active ingredient, Outcome 1 Presence of stretch marks.
Review:
Comparison: 2 Topical preparations with active ingredients compared with other topical preparations with active ingredient
Outcome: 1 Presence of stretch marks
Other active
preparations
Study or subgroup
Active preparations
n/N
n/N
Buchanan 2010
54/122
68/123
61.7 %
0.80 [ 0.62, 1.03 ]
3/30
12/30
38.3 %
0.25 [ 0.08, 0.80 ]
152
153
100.0 %
0.51 [ 0.16, 1.60 ]
de Buman 1987
Total (95% CI)
Risk Ratio
MH,Random,95%
CI
Weight
Risk Ratio
MH,Random,95%
CI
Total events: 57 (Active preparations), 80 (Other active preparations)
0.02
0.1
1
10
50
Favours other active prep
Analysis 2.2. Comparison 2 Topical preparations with active ingredients compared with other topical
preparations with active ingredient, Outcome 2 Severity of stretch marks.
Review:
Comparison: 2 Topical preparations with active ingredients compared with other topical preparations with active ingredient
Outcome: 2 Severity of stretch marks
Study or subgroup
Other active
preparations
Active preparations
Mean
Difference
N
Mean(SD)
N
Mean(SD)
Buchanan 2010
101
1 (1.2)
105
1.2 (1.2)
Total (95% CI)
101
Weight
IV,Random,95% CI
Mean
Difference
IV,Random,95% CI
100.0 %
-0.20 [ -0.53, 0.13 ]
100.0 % -0.20 [ -0.53, 0.13 ]
105
Heterogeneity: not applicable
-1
-0.5
0
0.5
1
Favours other active prep
32
APPENDICES
Appendix 1. Methods used to assess trials included in previous versions of this review
The following methods were used to assess Mallol 1991, Wierrani 1992 in previous versions of this review (Young 1996).
We evaluated trials under consideration for methodological quality and appropriateness for inclusion, without consideration of their
results. We processed trial data as described in Clarke 2000.
WHAT’S NEW
Last assessed as up-to-date: 6 March 2012.
Date
Event
Description
31 October 2011
New citation required and conclusions have changed
Two new authors helped prepare this updated review.
After restructuring the review’s comparisons the review
found no high-quality evidence to support the use of any
of the topical preparations in the prevention of stretch
marks during pregnancy
31 October 2011
New search has been performed
Two studies (three reports) identified in an earlier search
have now been included in the review (de Buman 1987;
Osman 2008).
An updated search identified a further six trials: two studies have been included (Buchanan 2010; Taavoni 2011)
; two have been excluded (Mendez Velarde 2010; Msika
2002) and two are awaiting classification (Lachmann
2011; Ortega 1985).
This review is now comprised of six included studies,
four excluded studies and two studies that are awaiting
classification
The title has changed from ’Creams for preventing
stretch marks in pregnancy’ and the methods have been
updated
33
HISTORY
Protocol first published: Issue 2, 1996
Review first published: Issue 2, 1996
Date
Event
Description
1 October 2009
Amended
Search updated. Three reports added to Studies awaiting classification
1 September 2008
Amended
Converted to new review format.
30 April 2004
New search has been performed
Search updated. A second study (Wierrani 1992) has been reviewed. This
compares massage using an ointment containing several possibly active ingredients with no treatment.
CONTRIBUTIONS OF AUTHORS
All review authors (MB, GY and DD) prepared this review update.
DECLARATIONS OF INTEREST
None known.
DIFFERENCES BETWEEN PROTOCOL AND REVIEW
We have separated outcomes into primary and secondary outcomes. The outcome from the previous version of this review (presence
of stretch marks) is our primary outcome and we have added a new secondary outcome (severity of stretch marks). The methods have
been updated to reflect the latest Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
We have also changed the title of the review from ’Creams for preventing stretch marks in pregnancy’ to ’Topical preparations for
preventing stretch marks in pregnancy’.
For this update we restructured the review comparisons to compare: (1) topical preparations with active ingredients compared with
placebo or no treatment, and (2) topical preparations with active ingredients compared with other topical preparations with active
ingredients. This is in contrast to comparing active creams with placebo or with no treatment, as presented in Young 1996.
NOTES
In the next update of this review we will carry out subgroup analysis by parity (nulliparous versus multiparous women), and our criteria
for sensitivity analysis will incorporate trials at high risk of bias for blinding. We will also detail how the primary (presence of stretch
marks) and secondary (severity of stretch marks) outcomes are measured and by whom. We will include a discussion around if and how
included studies have addressed confounding or other risk factors.
34
INDEX TERMS
Medical Subject Headings (MeSH)
Cosmetics; Dermatologic Agents [∗ administration & dosage]; Ointments; Randomized Controlled Trials as Topic; Skin; Striae Distensae
[∗ prevention & control]
MeSH check words
Female; Humans; Pregnancy
35

Topical preparations for preventing stretch marks in pregnancy (Review) The Cochrane Library

Transcription

Similar documents

Chain Drug Review - Dr. Tess Mauricio

Pattern Magic

INDEMNITY: EXPECTANT MOTHERS Please complete

Edexcel GCSE Art and Design Unit 2: Externally Set Assignment

pregnancy STRETCHES: TRY AND HOLD ALL STRETCHES FOR 15 – 20 SECONDS

Stretch Marks: The Facts

Triamcinolone acetonide ARISTOCORT R CREAM 0.1% ARISTOCORT C CREAM 0.5%

TRAVELODGE ROOM WORKOUT

EXPERIENCE TISHA B`AB AT BEVIS MARKS SYNAGOGUE